Error in returnAllowedChromsBaf -> returnBAlleleFreqs

[sunyihong]

Hello, when I run somatic CNV, I got this error:

[1] "We are reading BAF files. It may take time - especially if you have a lot of SNV positions. 2024-05-14 10:40:39"
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.4007 0.4545 0.4841 0.4856 0.5141 0.5988
Error in if (!is.na(potentiallyHeterozygous)) { :
the condition has length > 1
Calls: returnAllowedChromsBaf -> returnBAlleleFreqs
Execution halted

any suggestions? thank you

[GermanDemidov]

Hi, could you put a piece of code

print(potentiallyHeterozygous)

at the line 119 in this file

https://github.com/imgag/ClinCNV/blob/master/somatic/bafSegmentation.R

and send me the output?

It should be a simple vector and its summary seems to be sensible, but R somehow thinks it is not a simple vector, I guess...

[GermanDemidov]

No, don't do this, replace

!is.na(potentiallyHeterozygous)

with length(potentiallyHeterozygous ) > 0

and let me know if it works!

[sunyihong]

Hi, then I got the new error:
[1] "We have found sample with name A1_1_n in our normal.cov file! We will try to analyse it. Remember - we still have to calculate all the parameters (the most time consuming step)..."
[1] "We are reading BAF files. It may take time - especially if you have a lot of SNV positions. 2024-05-14 15:55:11"
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.4007 0.4545 0.4841 0.4856 0.5141 0.5988
[1] "Heterozygous allele shift for particular sample 0.484142"
Error in if (is.na(prob)) { : the condition has length > 1
Calls: returnAllowedChromsBaf ... apply -> FUN -> determineHeterozygousPositionsOverdispersed
Execution halted

[GermanDemidov]

in old R versions it was an acceptable code style but I think it fails with the newer ones...

is.na(prob) in https://github.com/imgag/ClinCNV/blob/master/somatic/helpersBalleleFreq.R has to be replaced any(is.na(prob)), in two places. Could you do it and tell me the next error, if any?

Which R version do you use? Because if it is the new one I probably have to adapt the rest of the script and make a new release.

[sunyihong]

The R version I used is 4.2.2, and I changed the code following your advice, got below error:

1] "We are reading BAF files. It may take time - especially if you have a lot of SNV positions. 2024-05-14 16:47:41"
Min. 1st Qu. Median Mean 3rd Qu. Max.
0.4007 0.4545 0.4841 0.4856 0.5141 0.5988
[1] "Heterozygous allele shift for particular sample 0.484142"
Error in if ((prob > 0.05)) { : the condition has length > 1
Calls: returnAllowedChromsBaf ... apply -> FUN -> determineHeterozygousPositionsOverdispersed
Execution halted

[sunyihong]

But I use R version 4.1.2 without any code change, it works!

[GermanDemidov]

Hi @sunyihong - so it works for you now?

I will debug the tool for the newest version, seems that something fundamentally changed in the language (so many conditions failed, damn), so I'll try to hunt it down, but probably not today

[sunyihong]

Yes! Thanks！！！